Pediatric acute myeloid leukemia remains a deadly disease with an overall survival of 53%. Recent treatment innovations have focused on adding targeted therapies to block key cell signaling pathways. However the pathways to target for each patient vary and the potential efficacy of targeted therapy is compromised by our inability to predict which agents will be effective for an individual patient. A greater understanding of the global changes that occur in protein activation following treatment could provide crucial information that would improve our knowledge gap and potentially increase the efficacy of targeted therapy. The Children's Oncology Group (COG) is conducting a phase 3 randomized clinical trial that will randomize 1050 patients to receive either standard chemotherapy (cytarabine, daunomycin, etoposide = ADE) or ADE with the proteasome inhibitor bortezomib (ADEB), a drug known to block protein degradation. We are assessing the activation states of key cellular proteins in patients enrolled in this trial before and during the first day of treatmet. Our long-term goal is to understand the importance of protein cell stress pathways in chemotherapy responsiveness, with the goal of manipulating cell stress protein pathways to increase chemotherapy effectiveness. The overall objective of this application is to determine if cell stress pathways such as the unfolded protein response (UPR) can identify proteins that predict therapy response. Our central hypothesis is that UPR activation, and activation of other cell stress pathways, can be used to predict ADE response. Two innovative approaches will be used to assess protein activation states: reverse-phase protein lysate arrays (RPPA) will be used to assess static protein changes before and during ADE+/-B treatment. Single cell networking profiles (SCNP) will be used to provide comprehensive functional assessments of signaling pathways in pre-treatment AML samples. Our central hypothesis will be tested with the following specific aims: 1) Determine if UPR activation, either at baseline or following chemotherapy, modulates chemotherapy response, 2) Determine if changes in cell stress protein activation, in both myeloblasts and in AML-leukemia initiating cells (LIC), predict response to ADE +/-B therapy, and 3) Use RPPA and SCNP to build and validate multi-variate classifiers that are a) predictive of ADE response, and b) are prognostic of clinical outcome measures such as relapse risk. Successful completion of the proposed research will have a significant impact in the field of AML biology for two reasons. First, this work will lay the foundation to add protein cell stress markers to the armamentarium of biomarkers currently used to assess relapse risk in AML. This will enhance risk stratification by identifying patients most likely to benefit from the addition of biologic agents to their treatment regimen. Second, our studies will also provide prospective validation of the clinical usefulness of the RPPA and SCNP techniques in the set- ting of a phase 3 clinical trial. Proteomic data from either RPPA or SCNP molecular diagnostics could lead to the development of key assays for response prediction in AML.